Characterizing Service Providers Traffic of Mobile Internet Services in Cellular Data Network

摘要

With the rapid rising of the number of cellular phone users, accessing the Internet via mobile device is becoming a new norm in terms of network activities and people's daily life, resulting in an ever-increasing usage of the mobile Internet. This paper analyze Service Provider (SP) traffic characteristics Based on the massive data collected from a commercial Internet Service Provider (ISP) covering an entire province in Southern China. We investigate the distribution of SP traffic volume and flow number. In particular, our study has scrutinized the rank of SP in terms of traffic volume distribution, flow number distribution, and subscriber number distribution, fitted with a distribution model named power law distribution. Furthermore, we characterize the correlations among the traffic volume, flow number and subscriber number. As distributed parallel processing system have been recently developed due to the cloud computing platform and the cluster file system, they could be usefully applied to analyzing big traffic data. Thus, in this paper, we propose a mobile Internet traffic analysis method Based on the Map Reduce software framework of the cloud computing platform for a large-scale network.